Atmos. Chem. Phys., 16, 13601–13618, 2016 www.atmos-chem-phys.net/16/13601/2016/ doi:10.5194/acp-16-13601-2016 © Author(s) 2016. CC Attribution 3.0 License.

Unexpectedly acidic nanoparticles formed in dimethylamine–ammonia–sulfuric-acid nucleation experiments at CLOUD

Michael J. Lawler1,a,b, Paul M. Winkler2, Jaeseok Kim3,4, Lars Ahlm5, Jasmin Tröstl6, Arnaud P. Praplan7,8, Siegfried Schobesberger7,9, Andreas Kürten10, Jasper Kirkby10,11, Federico Bianchi7, Jonathan Duplissy7, Armin Hansel12, Tuija Jokinen7, Helmi Keskinen3,7, Katrianne Lehtipalo7,6, Markus Leiminger12, Tuukka Petäjä7, Matti Rissanen7, Linda Rondo10, Mario Simon10, Mikko Sipilä7, Christina Williamson10,13,14, Daniela Wimmer7,10, Ilona Riipinen5, Annele Virtanen2, and James N. Smith1,b 1Department of Chemistry, University of California, Irvine, Irvine, CA, 92697, USA 2Faculty of Physics, University of Vienna, 1090 Vienna, Austria 3Department of Applied Physics, University of Eastern Finland, Kuopio, Finland 4Arctic Research Center, Korea Polar Research Institute, Yeonsu-gu, Incheon 21990, Republic of Korea 5Department of Environmental Science and Analytical Chemistry, Stockholm University, Stockholm, Sweden 6Paul Scherrer Institute, Villigen, Switzerland 7Department of Physics, University of Helsinki, 00014 Helsinki, Finland 8Finnish Meteorological Institute, 00101 Helsinki, Finland 9Department of Atmospheric Sciences, University of Washington, Seattle, WA 98195, USA 10Institute for Atmospheric and Environmental Sciences, Goethe University of Frankfurt, 60438 Frankfurt am Main, Germany 11European Organization for Nuclear Research (CERN), Geneva, Switzerland 12Institute for Ion and Applied Physics, University of Innsbruck, 6020 Innsbruck, Austria 13Cooperative Institute for Research in Environmental Sciences, University of Colorado Boulder, Boulder, CO, USA 14Chemical Sciences Division, NOAA Earth System Research Laboratory, Boulder, CO, USA avisitor at: National Center for Atmospheric Research, Atmospheric Chemistry Observations and Modeling Lab, Boulder, CO, 80301, USA bformerly at: University of Eastern Finland, Department of Applied Physics, Kuopio, Finland

Correspondence to: Michael J. Lawler ([email protected])

Received: 27 April 2016 – Published in Atmos. Chem. Phys. Discuss.: 24 June 2016 Revised: 16 September 2016 – Accepted: 29 September 2016 – Published: 3 November 2016

Abstract. New particle formation driven by acid–base chem- for the collected 10–30 nm VMD particles. This behavior istry was initiated in the CLOUD chamber at CERN by intro- is not consistent with present nanoparticle physicochemical ducing atmospherically relevant levels of gas-phase sulfuric models, which predict a higher dimethylaminium fraction acid and dimethylamine (DMA). Ammonia was also present when NH3 and DMA are present at similar gas-phase con- in the chamber as a gas-phase contaminant from earlier ex- centrations. Despite the presence in the gas phase of at least periments. The composition of particles with volume median 100 times higher base concentrations than sulfuric acid, the diameters (VMDs) as small as 10 nm was measured by the recently formed particles always had measured base : acid ra- Thermal Desorption Chemical Ionization Mass Spectrome- tios lower than 1 : 1. The lowest base fractions were found in ter (TDCIMS). Particulate ammonium-to-dimethylaminium particles below 15 nm VMD, with a strong size-dependent ratios were higher than the gas-phase ammonia-to-DMA ra- composition gradient. The reasons for the very acidic com- tios, suggesting preferential uptake of ammonia over DMA position remain uncertain, but a plausible explanation is that

Published by Copernicus Publications on behalf of the European Geosciences Union. 13602 M. J. Lawler et al.: Acidic nanoparticles at CLOUD the particles did not reach thermodynamic equilibrium with particles grow, and water vapor has been shown to be an im- respect to the bases due to rapid heterogeneous conversion of portant contributor to the initial steps of new particle forma- SO2 to sulfate. These results indicate that sulfuric acid does tion (e.g., Zollner et al., 2012, and references therein). While not require stabilization by ammonium or dimethylaminium ammonia is present at much higher concentrations than DMA as acid–base pairs in particles as small as 10 nm. in the remote atmosphere, amines such as DMA can effec- tively compete with ammonia in the growth of nanoparti- cles of ∼ 10 nm diameter (Smith et al., 2010; Barsanti et al., 2009). Measurements made at the CLOUD chamber in 1 Introduction CERN of molecular clusters in the sulfuric acid–base sys- tem show a nucleation process involving stepwise additions Atmospheric new particle formation (NPF) refers to the for- of H2SO4 and base, typically in a 1 : 1 ratio (Kirkby et al., mation and subsequent growth of condensed-phase particles 2011; Schobesberger et al., 2013, 2015; Almeida et al., 2013; from gas-phase precursors. The mechanisms and chemical Kürten et al., 2014; Bianchi et al., 2014). At this early stage species responsible for NPF have been the subject of many in the particle formation process, when there are fewer than recent studies, with the motivation of understanding the im- about 20 molecules in the clusters, ionic acid–base interac- pacts of these processes on air quality and climate. NPF is an tions clearly drive the growth. For these experiments, sulfuric important source of cloud condensation nuclei (CCN) and acid was present at much lower concentrations than were the therefore may significantly affect the global radiative en- bases, so it was the chemical species that limited the growth. ergy balance (Wang and Penner, 2009; Kazil et al., 2010), Here we present chemical composition measurements of but the mechanisms and chemical species driving NPF are recently formed nanoparticles in the size range of 10–30 nm still poorly understood (e.g., Riipinen et al., 2012). Sulfuric in diameter (volume median diameter for collected parti- acid is widely accepted as one of the most important species cles), from experiments conducted in the CLOUD chamber for particle formation in the atmosphere (e.g., Kirkby et al., at CERN. These are the first direct compositional measure- 2011; Kuang et al., 2008; Kulmala et al., 2007; Weber et ments of newly formed particles in this chemical system. al., 1997). However, particle growth rate and composition These observations place constraints on the nature of parti- measurements have indicated that species other than sulfu- cle growth after nucleation in the DMA–NH3–H2SO4–water ric acid dominate growth past the smallest particle sizes in system, with implications for new particle formation in the many environments (O’Dowd et al., 2002; Weber et al., 1997; atmosphere. Smith et al., 2008; Kuang et al., 2010; Bzdek at al., 2012). Three fundamental types of growth pathways for nanopar- ticles have been identified in the literature: reversible con- 2 Laboratory conditions densational growth by low-volatility species, reactive up- take, and acid–base interactions (Riipinen et al., 2012, and The experiments were undertaken at the seventh Cosmics references therein). Recent evidence has supported an im- Leaving Outdoor Droplets (CLOUD7) campaign at CERN portant role for highly oxidized, very low-volatility organic during October 2012. The CLOUD chamber is a 26.1 m3 molecules in the growth of atmospheric nanoparticles (Ehn cylindrical chamber made from electropolished stainless et al., 2014; Zhao et al., 2013; Riccobono et al., 2014). Reac- steel. Efforts were taken in its construction to make it the tive uptake involves the multiphase transformation of a gas- cleanest possible chamber used for studying atmospheric phase species into a new condensed-phase species that is less particle nucleation (Kirkby et al., 2011; Schnitzhofer et al., volatile than its precursor (e.g., Wang et al., 2010). Growth 2014; Duplissy et al., 2016). It can be precisely temperature due to acid–base chemistry results from strong ionic inter- controlled, and UV light at 250–400 nm can be provided to actions, which could be reversed depending on the chemical the chamber with minimal heating via an array of fiber-optic conditions within the cluster or particle (Bzdek et al., 2010; bundles (Kupc et al., 2011). It is possible to eliminate ions Barsanti et al., 2009). from the chamber using a 20 kV m−1 clearing field and con- In the sulfuric acid–dimethylamine–ammonia–water sys- versely to increase the number of chamber ions by directing a tem, the most likely growth pathways involve electrostatic 3.5 GeV pion beam into the chamber. The latter state results interactions arising from proton transfer from sulfuric acid in ion concentrations that are up to an order of magnitude to the base species, direct condensational growth by sulfuric greater than what is present naturally from cosmic ray bom- acid due to its very low saturation vapor pressure, and coag- bardment at the Earth’s surface (Franchin et al., 2015). ulation of particles and/or clusters. Dimethylamine (DMA) Clean air and reagents were added to the chamber on a and ammonia on their own are quite volatile gases, and sig- continuous basis, yielding a roughly 4 h overturning time for nificant growth by water vapor condensation alone is unfa- the chamber. The chamber conditions were set to simulate vorable due to the Kelvin effect until the particles are larger atmospheric conditions relevant to the lower troposphere. than about 50 nm in diameter. Nonetheless, all of these com- The main constituent gases were N2 (79 %) and O2 (21 %) pounds may be incorporated together in a stable matrix as the from cryogenic liquid sources. The chamber was maintained

Atmos. Chem. Phys., 16, 13601–13618, 2016 www.atmos-chem-phys.net/16/13601/2016/ M. J. Lawler et al.: Acidic nanoparticles at CLOUD 13603

Table 1. Overview of experimental conditions for the TDCIMS measurements described. The chamber temperature was 278 K and the relative humidity was 38 %. DMA and NH3 measurements are given (parts per trillion by volume) when coincident with the TDCIMS observations, and otherwise averages for the same experimental conditions close in time are given as estimates, indicated with parentheses. O3 and SO2 are reported in parts per billion by volume. Larger UV aperture fractions allow more ultraviolet light into the chamber, resulting in greater rates of SO2 oxidation to H2SO4.

Run no. DMA target DMA NH3 O3 SO2 UV aperture H2SO4 ppt ppt ppt ppb ppb % × 107 cm−3 1032 40 51 22 23 63 100 1.3–1.4 1033 40 (46) (28) 23 63 100 1.3 1035 40 (46) (28) 23 63 40 0.5–0.8 1036 40 (46) (28) 23 63 100 1.1–1.3 1040 10 (23) (19) 23 63 20 0.4 1043 10 (23) (19) 23 63 40 0.9–1.0 1047 10 28 28 86 63 100 2.4–2.7 1056 0 < 5 < 5 23 72 100 3.5 at 38 % relative humidity with ultrapure water generated by chromatography (IC), as described in Praplan et al. (2012). a Millipore filter system with UV irradiation (Merck Milli- The IC was not working during every experiment described pore). The trace gases added were ozone (O3), DMA, and here, but it provided measurements of both NH3 and DMA sulfur dioxide (SO2).O3 was usually kept at 23.5 parts per under several important experimental conditions. DMA was billion by volume (ppb), and DMA was set to either 10 or measured at an average mixing ratio of 46 ± 12 ppt (1 SD) 40 parts per trillion by volume (ppt) (see Table 1). SO2 for nominal 40 ppt DMA experiments, for which a total of was set to 63 ppb to facilitate the formation of H2SO4 at 18 particle collections are reported here, and 23 ± 7 ppt for atmospherically relevant rates, leading to levels from 5 to nominal 10 ppt DMA experiments, for which a total of 21 6 −3 35 × 10 cm . SO2 was the only species added at levels well particle collections are reported here (Simon et al., 2016). above likely background atmospheric mixing ratios. The ad- Over the respective periods, the average NH3 was 28 ± 9 dition of UV light resulted in OH radical formation, causing and 19 ± 10 ppt. Therefore in both the high and low DMA SO2 oxidation to H2SO4 and formation of new particles. The cases, the DMA and NH3 mixing ratios were comparable. It ozone mixing ratio was increased to 86 ppb for one experi- is worth noting that the IC is likely to measure particulate am- ment to increase H2SO4 levels and the new particle forma- monium and dimethylaminium as well as gas-phase ammo- tion rate. During CLOUD7 the chamber air was maintained nia and dimethylamine. During the high-temperature clean- at 278 K. Aerosol cleaning cycles were performed regularly ing, ammonia reached extremely high levels (over 1000 ppt) between experiments. For these cleanings, the chamber was due to evaporation from the chamber walls, whereas DMA flushed with clean humidified air and the particles were ion- only reached about 30 ppt. After the cleaning, both species ized with a pion beam generated from the CERN Proton Syn- were below the 5 ppt detection limit for the experiment with chrotron, with the high voltage clearing field being cycled no added DMA. on and off to allow charging and then removal of charged Many instruments sampled the chamber air to assess the particles and clusters. An extended cleaning of the chamber characteristics of the gases, molecular clusters, and particles was performed prior to the campaign. This involved rinsing present in the chamber. These included a scanning mobil- the chamber walls with ultrapure water for 24 h, followed by ity particle sizer (SMPS), a hygroscopicity tandem differ- raising the temperature to 373 K for a further 24 h and flush- ential mobility analyzer (HTDMA; Kim et al., 2016; Ke- ing with ultrapure synthetic air to drive contaminants off the skinen et al., 2013), and several mass spectrometers. Atmo- chamber walls. A further overnight heating cycle at 373 K spheric pressure interface time-of-flight mass spectrometers was performed prior to the experiment with no added DMA (APi-TOFs, Tofwerk AG) measured gas-phase ions (Kirkby (the final experiment described here). et al., 2011; Schobesberger et al., 2013; Almeida et al., 2013; Despite the efforts to minimize contamination, ammonia Bianchi et al., 2014), and two chemical ionization APi-TOFs was present as an impurity during the DMA experiments measured neutral gas-phase species (Jokinen et al., 2012; as a carryover from previous experiments during the cam- Kürten et al., 2014). paign. A few days prior to the DMA–sulfuric acid experi- ments described in this study, nucleation experiments with intentional addition of ammonia gas at levels of 20 and 40 ppt were performed. After that time, no ammonia was inten- tionally added. However, ammonia was detected using ion www.atmos-chem-phys.net/16/13601/2016/ Atmos. Chem. Phys., 16, 13601–13618, 2016 13604 M. J. Lawler et al.: Acidic nanoparticles at CLOUD

3 TDCIMS instrument CLOUD chamber Ar MFC N2 The thermal desorption chemical ionization mass spectrome- +H2 ter (TDCIMS) has been described in detail elsewhere (Smith MFC et al., 2004; Lawler et al., 2014). Significant improvements 210Po 210Po MFC in ammonium sensitivity were made since the measure- 210 ments described in Lawler et al. (2014) by modifying the to Po Mass vac differential pumping and ion transfer optics. For these ex- UPCs spectrometer periments, particles were sampled continuously from the MFC −1 CLOUD chamber at 3.3 L min (Fig. 1). Sampled particles Collection were charged with a pair of unipolar chargers (UPCs; Mc- RDMAs filament

Murry et al., 2009; Chen and Pui, 1999). The charged par- CPC ticles were then collected by electrostatic deposition onto a Pt filament. After a collection period of typically 30 min, the filament was translated into the ion source of the mass Recirc. sheaths spectrometer and the temperature was ramped to ∼ 600 ◦C to SMPS MFC to to vac desorb the collected mass. Desorbed molecules and decom- vac position products were ionized and ultimately detected using a high-resolution time-of-flight mass spectrometer (HTOF, Figure 1. Schematic of the TDCIMS instrument as deployed at Tofwerk AG). Instrument backgrounds were assessed for CLOUD7. Dashed boxes show components that were not always connected. Particles from sampled chamber air were negatively each collection by performing the identical procedure but charged in the UPCs (unipolar chargers), optionally size-selected without applying a collection high voltage. All measure- by radial differential mobility analyzers (RDMAs), and electrostat- ments reported here are corrected for this background. The ically precipitated onto a Pt collection filament (shown here in col- TDCIMS is capable of detecting both positive and negative lection position) held at +4 kV over a period of 15 or 30 min. The ions, but only one polarity can be monitored for each sample filament was then translated vertically into the ion source of a time- collected. Chemical ionization reagent ions are generated by of-flight (TOF) mass spectrometer and the sampled material was 241 an Am source kept in a clean N2 flow. Impurity H2O and thermally desorbed and/or decomposed by a temperature ramp of + O2 in the N2 gas result in (H2O)nH reagent ions for positive the wire at atmospheric pressure in a dry N2 environment. The − resulting compounds were ionized and passed into the TOF via mode and (H2O)nO2 reagent ions in negative mode. Particle- + + electrostatic and octopole ion guides. CPC: condensation particle phase dimethylaminium (C2H8N , or DMAH ) and ammo- nium (NH+) are detected as molecular ions and as ions clus- counter. MFC: mass flow controller. Vacuum pumps and vacuum 4 lines have been omitted from the figure. tered with water in positive mode. Sulfate is detected in neg- − ative ion mode primarily as the SO5 ion, which is formed from the reaction of the sulfate salt decomposition product − SO3 with O2 . The TDCIMS observations are presented as precipitator captured sampled particles. Additionally, a post- sums of background-corrected detected ions, integrated over campaign test was performed using a long-column differen- the desorption period, for each collected sample, or as ratios tial mobility analyzer (TSI Inc., model 3081; long DMA) of these ion sums. to assess the capture efficiency of 12–140 nm particles un- During most of the CLOUD7 experiments, the sampled der the same instrumental operating conditions. The region particles were not mobility-selected, and the particle sizes of overlap between these two collection efficiency mea- that made up most of the collected mass were dependent on surements compared well. The two efficiency measurements the chamber particle size distribution as it evolved over the were combined for correcting the experimental data, and the course of the particle formation events. For one experiment, absolute cutoff for collection was determined to be about a mobility size of 20 nm was selected using a pair of radial 60 nm (Fig. 2). Charged particles smaller than about 15 nm differential mobility analyzers (RDMAs; Zhang et al., 1995) appeared to be captured with ∼ 100 % efficiency once inside placed between the unipolar chargers and the electrostatic the precipitator region. precipitator. This mostly served to exclude small (< 15 nm) To assess the net size-resolved collection efficiency of particles from analysis. A nano-SMPS sampled particles af- the instrument, sampling losses and (multiple) charging ef- ter charging in the UPCs. The nano-SMPS consisted of a ficiency had to be taken into account. This was achieved by a differential mobility analyzer (TSI Inc., model 3085; Nano statistical optimization procedure that relied on the CLOUD DMA) and an ultrafine condensation particle counter (CPC; chamber SMPS system and a CPC downstream of the TD- TSI Inc., model 3027) and was capable of measuring parti- CIMS precipitator, using a procedure similar to that de- cles of 4–42 nm mobility diameter. The nano-SMPS system scribed in Lawler et al. (2014). Briefly, a size-resolved sam- was temporarily moved downstream of the electrostatic pre- pling and charging efficiency filter was optimized to obtain cipitator to assess the size-resolved efficiency with which the agreement between the chamber particle distribution and the

Atmos. Chem. Phys., 16, 13601–13618, 2016 www.atmos-chem-phys.net/16/13601/2016/ Cal

M. J. Lawler et al.: Acidic nanoparticles at CLOUD 13605

5 5 0.20 4x10 (a) 4x10 (b)

0.15 3 3

0.10 2 2

0.05 1 1

N signal base

Sulfate signal

Net eff. collection 0.00 0 0 5 15 25 35 45 55 65 0.0 1.0 2.0 3.0 0.0 1.0 2.0 3.0 Particle diameter (nm) -9 3 -9 3 Collected particle volume (10 cm ) Collected particle volume (10 cm ) Figure 2. Solid line: estimated TDCIMS particle sampling ef- Figure 3. (a) Sum of oxidized sulfur (“sulfate”) signal vs. collected ficiency as a function of mobility diameter for bulk (non-size- + particle volume. (b) Sum of N base (DMAH+ and NH ) signal vs. resolved) sampling. Dashed line: the same number efficiency scaled 4 collected particle volume. Red crosses are from chamber particle by the volume of an individual particle at each size and normalized collections of DMA + H SO nucleation experiments, and black to fit on the same plot (arbitrary units). These estimates include the 2 4 circles are from calibration ammonium sulfate aerosol. Ion counting effects of sampling losses, (multiple) charging efficiency, and effi- errors are smaller than the symbols and are excluded. The negative ciency of capture by the electrostatic precipitator. Small (< 15 nm) ion data from the chamber particles are broadly consistent with par- particles are efficiently captured due to their high mobility. The effi- ticles dominated in volume by sulfate. However, the chamber par- ciency of collection at sizes below 4 nm could not be assessed using ticles have a much lower base fraction than the ammonium sulfate the available experimental apparatus, but particles at smaller sizes calibration aerosol. are not likely to have contributed detectable amounts of particle mass during these experiments.

within the charger (see Sect. 4). Adding the H2 gas resulted in higher background signals and variability for many of the TDCIMS CPC when there was no precipitator voltage ap- signals due to contaminants in the H / Ar gas cylinder. plied and after correcting for UPC-generated particles (see 2 next paragraph). This information was combined with the 3.1 Instrument calibrations precipitator capture efficiency to determine the sizes and total volume of particles collected. The volume median diameter 3.1.1 Calibrations during the CLOUD7 measurements (VMD) was assessed as a measure of the center of the col- lected particle distributions. The TDCIMS was calibrated for ammonium sulfate (AS) When sampling chamber air with the TDCIMS, there was particles once during the campaign by using an atomizer to a persistent mode of small (∼ 6 nm) particles generated in the generate aerosol from a dilute AS solution. The wet parti- UPCs. This was a result of ionization and oxidation processes cles were passed through a diffusion dryer before entering driven by the UPC radioactive sources. The particles were the UPCs. This calibration aerosol was mobility-selected to present even in the absence of H2SO4 in the chamber, indi- either 15 or 20 nm using a pair of RDMAs, one downstream cating that oxidation of SO2 to form H2SO4 was probably of each UPC, to control the amount of mass available for col- involved. The alpha-emitting radioactive sources are known lection on the filament. By selecting for 15 nm particles and − to produce OH radicals from H2O present in the sampled air collecting for 5 min, about 0.3 × 10 9 cm3 of particle volume (e.g., He and Hopke, 1995). When these charger-produced − + was collected. This calibration showed SO5 and (H2O)NH4 particles were the only source of particles to the TDCIMS, signals in the low-to-mid range of those observed in the nu- the resultant background-corrected particle-phase ion signals cleation experiments (Figs. 3 and 4). By selecting 20 nm were usually not detectable, indicating that these fine parti- particles, an estimated particle volume of 1.3 × 10−9 cm3 cles did not contribute significantly to the TDCIMS cham- was collected, resulting in sulfate signals that were similar ber measurements. However, an attempt was made to try to to mid-to-large chamber aerosol samples and ammonia sig- reduce any possible impact of this oxidation process, as it nals over twice as large as any observed for chamber aerosol also seemed possible that charger-generated H2SO4 could in- (Figs. 3 and 4). These AS aerosol calibrations indicate that fluence the composition of sampled particles. During many the TDCIMS was roughly equally sensitive on a molar ba- of the experiments, a flow of 5 % H2 in Ar (industrial-grade sis to ammonium and sulfate during CLOUD7, with a 2.3 : 1 welding gas) was added to the TDCIMS sampling inlet to ammonium-to-sulfate-signal ratio for AS. achieve a ∼ 1 % H2 mixing ratio. This was done to scavenge OH and other oxidants generated in the UPCs and resulted 3.1.2 Post-CLOUD7 calibrations in a reduction in the spurious small particle mode by a factor of 3–4. Measured particle sulfate fractions did not appear to Aerosol calibrations were performed after the CLOUD7 ex- be substantially affected by this change (see Sect. 4). How- periments to assess the relative sensitivity of the TDCIMS ever, this test demonstrated that n-methyl formamide (MFA) to particle ammonium and dimethylaminium and to assess detected in the particles was the result of oxidation processes the linearity of the response. Aqueous solutions of 2 mM www.atmos-chem-phys.net/16/13601/2016/ Atmos. Chem. Phys., 16, 13601–13618, 2016 13606 M. J. Lawler et al.: Acidic nanoparticles at CLOUD

a. to characterize the resulting size distribution. During particle 3

120x10 SO5 96 collection, the nano-differential mobility analyzer was by- passed so that the number concentration difference between 100 sample and background could be used to assess the num-

80 ber of particles collected for brief collection periods. Sample time was varied to achieve different collection masses, which 60 were estimated using the known size distribution, the number

Ion Counts concentrations downstream of the wire during collections, 40 NO2 and the wire collection efficiency, which was determined us- SO4 ing a comparison of back-to-back stable collection and back- 20

NO2(H2O)

SO3 ground particle concentrations. For these calibrations, the dry

CHNO3

C8H4O3 [phthalic anhydride]

CHNO2

C2HO4 CNO [oxalic acid]

HSO4 0 particle densities were assumed to correspond to dimethy- −3 −3 40 50 60 70 80 90 100 110 120 130 140 150 lamine sulfate (1.35 g cm ) and AS (1.78 g cm ), respec- Atomic Mass (Da) tively (Qiu and Zhang, 2012), and collected masses are re- b. ported. It should be noted that the actual base : acid ratio of 3 250x10 89 the calibration aerosol is not known exactly because of loss

NH4(H2O) to the gas phase during evaporation. For example, Wong et 200 al. (2015) found that their wet AS aerosol had an ammo- nium : sulfate ratio of 1.72 due to this process. In addition, a 150 given salt may only have certain stable configurations when dried. A dimethylamine sulfate droplet was shown to go from Ion Counts 100 + 2− a 2 : 1 solution to a 1.5 : 1 DMAH : SO4 ratio particle when

NH4(NH3) dried (Chan and Chan, 2013). Ouyang et al. (2015) showed 50 a tendency for dried DMA : H SO nanoclusters to reach a

NH4 NH4(H2O)2 2 4 −

NO(H2O) 3

H7NO2

NO

C2H6NO [amide]

C2H6NO(H2O) [amide] consistent density of 1.567 g cm , independent of the bulk

H5NO 0 composition of the initial electrospray solution. Despite us- 20 30 40 50 60 70 80 ing solution DMA : H SO molar ratios ranging from 1 : 10 Atomic Mass (Da) 2 4 to 2 : 1, the particles settled on the same final composition. Figure 4. Calibration ammonium sulfate aerosol mass spectrum Similarly, during “droplet tests” in our lab in which dilute − − for (a) negative and (b) positive ions. SO5 and SO4 are the main salt solutions are directly applied to the TDCIMS collection − sulfate ions, and NO2 indicates a nitrate contaminant in the stan- wire, allowed to evaporate, then analyzed normally, we found dard, most likely introduced by the water or air used for the at- only a few percent increase in DMAH+ signal when using a omization. The instrument is more sensitive to ammonium nitrate + 2− − 2 : 1 DMAH : SO solution compared with a 1 : 1 solution, (detected as NO ) than to ammonium sulfate by a factor of ∼ 100 4 2 indicating that the additional DMA simply evaporated rather (Bzdek et al., 2014), so the nitrate contamination is likely a small + than being incorporated into the salt (the molar ratio of which fraction of the sampled aerosol mass. NH and its clusters with wa- 4 remains uncertain). ter and NH3 are the main ammonium ions. All species plotted have one elemental charge. The sensitivity of the TDCIMS to sulfate in the negative ion mode was basically indistinguishable for the two salt types (Fig. 5). The signal dependence was nonlinear with respect to sampled mass and the relationship is well-fit by AS and dimethylaminium bisulfate (DMABS) were atom- a second-order polynomial. Similar to the negative ion sig- ized in a stainless steel atomizer, dried by passing through nal, the base signals showed nonlinear sensitivity with re- a thermal denuder at 373 K, and sampled by TDCIMS. All spect to mass when a large range of collected mass was transport, sheath, and atomizer gas flows used N2 gas deliv- considered, and the relationship could be described using a + ered from a liquid N2 dewar. Tests in our lab have shown second-order polynomial. NH4 was detected as a contam- that even brief exposure to lab air or even “clean” zero inant in the DMABS aerosols. Contamination of the parti- air can cause significant contamination of the particles with cles by ammonia most likely occurred sometime prior to col- ammonium. The AS solution was generated using AS salt lection on the wire (i.e., either during generation or trans- (> 99 %, Sigma-Aldrich) and deionized water (Millipore). port of the particles), based on tests of temporal stability of The dimethylamine bisulfate solution was made using 40 % the particle composition after precipitation onto the collec- (by mass) dimethylamine in H2O (Sigma-Aldrich), H2SO4 tion wire. The solutions themselves showed very little am- (95–98 %, Sigma-Aldrich), and deionized water (Millipore) monium contamination, based on a test in which a dilute so- to form a 1 : 1 molar ratio of dimethylaminium to sulfate. The lution was directly applied to the precipitator wire. This test TDCIMS RDMAs were used to size-select the sampled parti- also indicated that there was no significant ammonia released cles at 25 nm mobility diameter, and a nano-SMPS was used upon thermal decomposition of the dimethylaminium salt.

Atmos. Chem. Phys., 16, 13601–13618, 2016 www.atmos-chem-phys.net/16/13601/2016/ M. J. Lawler et al.: Acidic nanoparticles at CLOUD 13607

(a) (b)

3 + 12 x10 4000 Amm. Sulf: NH4 Amm. Sulf: Sulfate Signal + DMA Bisulf: Sulfate Signal DMA Bisulf: DMAH 10 + DMA Bisulf: NH4 3000 + + 8 DMA Bisulf: DMAH + NH4

6 2000

4 1000 Positive ionsPositive counted Negative ions Negative counted 2

0 0 0.0 0.5 1.0 1.5 2.0 0.0 0.5 1.0 1.5 2.0 2.5 Collected particle mass [ng] Collected particle mass [ng]

Figure 5. Results of a post-CLOUD7 calibration of the TDCIMS with salts generated by atomizing liquid solutions of ammonium sulfate, (NH4)2SO4, and dimethylaminium bisulfate (DMABS), (C2H8N)HSO4. Particles of roughly 25 nm volume median diameter were used for this calibration, and larger mass collections were achieved by collecting for longer periods. For these calibrations, particle volume was corrected to mass using the densities of DMAS (1.35 g cm−3) and AS (1.78 g cm−3). (a) Signal in negative ion mode plotted against particle mass, with a second-order polynomial fit. The sulfate sensitivity for the two salts appears to be indistinguishable. (b) Signal in positive ion mode for the two salts. There was contamination of the DMABS by ammonium at some point between solution generation and analysis. Based on laboratory tests, this most likely occurs as a result of trace ammonia present in the carrier N2 and transport lines. For this reason, + + the DMAH signal, NH4 signal, and their sum are all given. The sum signal for DMABS is very similar to the AS signal, suggesting that both aerosols may actually have similar base : acid ratios after atomization and drying.

These observations underscore the facility with which am- usual way, with desorption proceeding a few seconds after + monia is taken up into DMA–H2SO4 aerosol. The DMAH the conclusion of sampling, and with an additional “rest” pe- signals for the DMABS calibration particles were slightly riod three times as long as the sampling period before des- + less than but comparable to the NH4 signals for the AS cal- orption to allow any volatile compounds to desorb. There was ibration particles. If the DMABS calibration aerosol was ex- no significant difference in the results from the two methods, + 2− actly 1 : 1 DMAH : SO4 and the AS aerosol was exactly indicating that, for these particles at least, continuous evapo- + 2− ration of base was not an issue. This test does not exclude the 2 : 1 NH4 : SO4 , this would indicate higher sensitivity to + + possibility that some DMA is lost rapidly during sampling, DMAH than to NH4 such that the reported CLOUD7 ex- + + i.e., as the result of the evaporation of an aqueous phase. perimental DMAH : NH4 ratios are overestimates. In this worst case, the DMAH+ sensitivity is about 50 % higher than However, the calibrations performed with atomized solutions + indicate that (DMA)H2SO4 and (NH3)2H2SO4 aerosol be- the NH4 sensitivity. However, based on the discussion pre- sented above, the actual composition ratios were probably have similarly in this regard. closer for the two calibration aerosols, and any inferred sen- sitivity difference would be smaller. If the actual composi- 3.1.3 Overview of calibrations tion ratios for the AS and DMABS aerosols were 1.75 : 1 + − − To summarize the calibration results, the TDCIMS shows NH : SO2 and 1.25 : 1 DMAH+ : SO2 , respectively, the 4 4 4 identical sensitivity to the sulfate fraction of the two cali- inferred sensitivity to both bases would be the same. The bration aerosols tested, DMABS and AS. However, there is base : acid signal ratios were roughly linear over a wide range some uncertainty in the relative sensitivity of the basic com- of collected masses (Fig. 6), allowing a straightforward anal- ponents NH+ and DMAH+. If there is indeed any differ- ysis of this quantity for the chamber aerosol. Overall, these 4 ence, the instrument is more sensitive to DMAH+ and there- calibrations indicate no significant sensitivity bias for the fore may overestimate both DMAH+ : NH+ and base : acid salts studied, but there is uncertainty in the relative sensitivity 4 + ratios. However, the measured experimental ratios were in of DMAH+ and NH due to uncertainty about the calibra- 4 fact lower than expected (see Sect. 4). The AS calibration tion aerosol composition. performed at CLOUD7 should be considered the best as- An additional test was performed after the CLOUD7 sessment of the relative sensitivities between positive (base experiments to determine the stability of DMA–H SO 2 4 ions) and negative (sulfate ions) ion modes for these mea- nanoparticles on the TDCIMS filament. Nanoparticles were surements because it was performed under identical operat- formed using high concentrations of DMA and H SO in a 2 4 ing conditions. This calibration showed the TDCIMS to be glass flow reactor modeled after that described in Glasoe et slightly more sensitive towards particulate ammonium than al. (2015). Particles of ∼ 5–50 nm were generated and sam- towards particulate sulfate, but given the very low base : acid pled with the TDCIMS. Particles were analyzed both in the ratios measured for chamber aerosol (see Sect. 4 below), no www.atmos-chem-phys.net/16/13601/2016/ Atmos. Chem. Phys., 16, 13601–13618, 2016 13608 M. J. Lawler et al.: Acidic nanoparticles at CLOUD

3 4000 a. 70x10

9 SO5 Ammonium sulfate 60

50 3000

40

30

2000 Ion Counts 20

H4O4

Base ionsSignal

SO4

10 CHNO2

NO2

NO2(H2O)

C3H3O3 [oxo-propanoic acid]

C8H4O3 [phthalic anhydride]

CHO2

HSO4 1000 SO3 0

40 50 60 70 80 90 100 110 120 130 140 150 Atomic Mass (Da) 0 b. 3 5000 0 2 4 6 8 10 12 14 x10 8 Acid ion signal

NH4(H2O) 4000 [ammonia]

C2H6NO 2500 [n-methyl formamide] 3000

C2H8N [DMA] C2H8N

NO(H2O)

C2H5NO(H3O) DMA bisulfate [n-methyl formamide]

C2H5O2(H2O) 2000 2000 [acetic acid]

Ion Counts

C2H8N(H2O) [DMA]

1000 H6O2S

CH5N2O

NH4 1500 [ammonia]

C2H4N

C2H6N C5H9O3

C8H15O2

C2H3N3O3 C7H15N2

C6H9S

C6H7O2 C7H13O C4H11O4

C8H9O2 C7H11O3

C6H13OS

C5H11N2O2 C6H9O4

C7H11O2

C3H7O5

C6H9O3

C8H11O2

C5H8NO4

C10H17

C8H19O

H9O4S 0

1000 20 30 40 50 60 70 80 90 Base Base ion signal Atomic Mass (Da)

500 Figure 7. Negative (a) and positive (b) background-corrected 30 min collection particle mass spectra for a particle forma- 0 tion experiment with nominal 40 ppt DMA, 23.5 ppb O3, and 6 −3 −3 0 2000 4000 6000 8000 2.8 × 10 cm cm H2SO4, run 1033 on 16–17 October 2012. Acid ion signal All species plotted have one elemental charge. The volume median + diameter of collected particles was 10.9 nm for positive ions and Figure 6. Base (DMAH+ + NH ) ion signals plotted against sul- 4 13.1 nm for the negative ions. Particulate ammonium and dimethy- fate ion signals for ammonium sulfate and dimethylaminium bisul- laminium levels were similar, despite not adding gas-phase ammo- fate calibration aerosol overestimated sample loadings of 0 to nia intentionally to the chamber. N-methyl formamide was a con- 1.8 ng, conducted in the laboratory at NCAR after CLOUD7 (the taminant generated in the particle chargers. The negative spectrum same tests as presented in Fig. 5). Signals were scaled by collected − is dominated by the sulfate-derived peak SO5 . The instrument is mass to match positive and negative analyses. Despite the nonlinear − character in the individual ions’ dependence on the sample mass more sensitive to ammonium nitrate (detected as NO2 ) than to am- ∼ (Fig. 5), the base : acid ratio remains constant over a range of col- monium sulfate by a factor of 100 (Bzdek et al., 2014), so the ni- lected masses for the same calibration aerosol, in agreement with trate contamination is likely a small fraction of the sampled aerosol the ammonium sulfate aerosol calibration performed at CLOUD7 mass. (Fig. 5). Note that relative sensitivities to base and acid were differ- ent during this test than during the calibrations at CLOUD7 (Fig. 5), − which should be considered the better comparison for the exper- HSO4 . The oxidized sulfur signal per estimated collected imental particle data since they were conducted under CLOUD7 mass is consistent with the AS calibration aerosol, indicat- conditions. ing that sulfate dominated the collected mass in the particles measured from the chamber, as expected (Fig. 3). Overall the spectra were very clean, but some apparent contaminant or- attempt was made to correct for this, and the base : acid ratios ganic molecules were also detected in the particles, including − − presented are strictly the ratios of the uncorrected ion signals. C3H3O3 (an oxidized carboxylic acid or fragment), CHNO2 − (likely fragment ion), and C8H4O3 (phthalic anhydride). Ph- thalic anhydride is most likely derived from phthalate con- 4 Results tamination by plastics. Dimethylaminium and ammonium were major compo- During the CLOUD7 experiments, negative ion TDCIMS nents of the positive ion spectra (Fig. 7b). The main ammo- − + spectra were dominated by SO5 (Fig. 7a). This was also the nium peak was its first water cluster, NH4 (H2O), which is strongest signal for AS in the calibrations (Fig. 4a) and re- also the main ammonium peak in the AS standards (Fig. 4b). + − + sults from the reaction of SO3 O2 in the ion source fol- There was also a large C2H6NO peak, which appears to lowing the decomposition of sulfate salts. There were other be the DMA oxidation product MFA, a contaminant gener- − − − sulfur masses that tracked SO5 , including SO3 , SO4 , and ated in the UPCs (see below and Fig. 8). In all experiments,

Atmos. Chem. Phys., 16, 13601–13618, 2016 www.atmos-chem-phys.net/16/13601/2016/ M. J. Lawler et al.: Acidic nanoparticles at CLOUD 13609

(a) 1.2 No H2

5 10 1.0

4 0.8 10

atio

r

+

0.6 3 10

+

Sum of ion counts of Sum SO5- DMA(H+) n-methyl formamide(H+) DMAHNH4 : 0.4 2 10 NH3(H3O+)

0.2

(b) 6 5 0.0 0 1 2 3 4 ng -9 3 Collected particle volume (10 cm ) collected 3

1 + + 2 Figure 9. The ratio of summed DMAH to summed NH4 ion signals plotted against the collected particle volume for 0.1 (nm)Diameter DMA + H2SO4 particle formation experiments (circles) and for the 10 0.01 9 H2SO4 only (no added DMA) nucleation experiment (triangle). 8 − − 7 Collections of less than 0.1 × 10 9 cm 3 of sample were excluded 0.001 6 5 due to low signal-to-noise ratio. The filled symbols are periods

4 0.0001 with nominally 10 ppt DMA in the chamber and the open symbols 12:00 PM 03:00 PM 06:00 PM 09:00 PM 12:00 AM are with 40 ppt DMA. Blue symbols indicate periods with 1 % H2 10/23/12 10/24/12 added to the TDCIMS inlet to scavenge radicals, and red symbols − are periods without H . The vertical error bars indicate the stan- Figure 8. (a) Particulate ion signals for sulfate (SO ), dimethy- 2 5 dard error of the ion signals. Higher ammonium fractions were more lamine, n-methyl formamide, and ammonia during a nucleation ex- likely for low collected masses (resulting from low chamber aerosol periment with 10 ppt DMA and 2.6 × 107 cm−3 H SO (run 1047). 2 4 mass), possibly because less contaminant ammonia was required to For most of this run, 1 % H was added to the TDCIMS inlet as 2 alter the composition of the particles (see Sect. 4). DMAH+ was a gas-phase radical scavenger, except for the period between the + never observed to be more abundant than NH in the particles dur- dashed vertical lines. During this period, the n-methyl formamide 4 ing any of experiments. increased about 30-fold, while the other species underwent more modest changes. (b) Size-resolved particle mass collected during this period. White gaps indicate background periods when no par- ticles were collected. Open circles show the volume mean diameter experimental condition was maintained for a period of over for collected particles. 3 h during the nucleation experiment, such that two measure- ments under “high charger oxidant” conditions were made for both positive and negative ion modes. Most of the main more ammonium was detected in the particles than dimethy- species observed were only minimally affected by this test, + laminium and no clear difference between nominal 40 and but C2H6NO in the particles increased by over an order of + + + 10 ppt DMA runs was observed (Fig. 9). The DMAH : NH4 magnitude (Fig. 8). The DMAH signal was not strongly ratio was more likely to be lower for smaller amounts of col- affected, indicating that DMA present in the sampled par- lected mass (Fig. 9), though a clear dependence on particle ticles was not appreciably oxidized in the absence of the H2 size was not observed. The collected mass dependence could scavenger. This suggests that the DMA oxidation to MFA be due to a limitation of the available contaminant ammo- occurred in the gas phase and the MFA partitioned to the nia or the aging time necessary to change the composition. particles afterward. It is also possible that the instrument Small mass collections were often the result of sampling at sensitivity for MFA is much greater than for DMAH+ and the end of runs, when the remaining particles were few, large, only a small fraction of particle-phase DMAH+ was oxi- and aged, with no more sub-10 nm particles present and col- dized. In either case, the large MFA signal is clearly spurious lected VMDs closer to 30 nm. In the run conducted with no and we have no evidence suggesting that the particle-phase + + DMA added, the DMAH signal was about 1 % of the NH4 DMA signal was significantly affected by the MFA produc- signal, indicating that DMA did not remain as a significant tion. MFA signal was therefore excluded from calculations of + + contaminant in the chamber, unlike NH3. base : acid and DMAH : NH4 ratios. Another concern was + To investigate the apparent spurious MFA (C2H6NO ) that other species, especially H2SO4, could be generated in formation in the UPCs, a test was performed in which the the chargers and partition to the particles. The addition of H2 oxidant-scavenging H2 flow to the TDCIMS inlet was turned reduced the fraction of charger-generated OH that could react off midway through a particle nucleation experiment. This with SO2 by almost 2 orders of magnitude. However, no con- www.atmos-chem-phys.net/16/13601/2016/ Atmos. Chem. Phys., 16, 13601–13618, 2016 13610 M. J. Lawler et al.: Acidic nanoparticles at CLOUD

2 dependence of 10 nm particle hygroscopicity on sulfuric acid but did observe a decrease in base : acid ratio for 15 nm par- 1 ticles as gas-phase sulfuric acid increased. For essentially all 9 8 7 collections, the TDCIMS base : acid ratio was below 0.5, in- 6

5 dicating that the particles did not reach a fully neutralized 4 state. For some collections of small (∼ 10 nm VMD) parti- 3 cles, this ratio was near 0.1, though there was high variabil-

2 Base : Acid Ratio ity for this size range. The HTDMA data show a similar size dependence but indicate higher base : acid ratios (Fig. 10). A

0.1 notable exception to the very acidic particles was a case for 9 8 which particles were mobility-selected at 20 nm before col- 7 6 lection and the base : acid ratio was 1.23. This is most likely 5 0 10 20 30 40 50 due to contamination of the particles by ammonia in the re- Particle diameter (nm) circulating sheath flow of the mobility analyzers and so the results of this experiment are therefore excluded from Figs. 9 Figure 10. Base : acid TDCIMS signal ratios for DMA–H SO ex- 2 4 and 10. The DMAH+ : NH+ ratio was low (∼ 0.15) for this periments (circles) and the H2SO4 alone experiment (triangle) plot- 4 ted against the volume median particle diameter for collected par- experiment, indicating a substantial change in particle com- ticles; signal-weighted APi-TOF positive ion cluster composition position relative to other experiments. The mass of small par- (black X markers) (Bianchi et al., 2014); HTDMA-based estimates ticles excluded during this test was too small to account for (black squares) assuming 1 : 1 ammonium : dimethylaminium ratio, the large change in the base : acid ratio, even if the excluded with standard deviations (n = 3–6) from Ahlm et al. (2016) and particles contained no DMA or ammonia. It should be noted Kim et al. (2016). The TDCIMS base : acid ratio is the summed that the same mobility analyzers were used to size-select the + + NH4 and DMAH ions and their clusters divided by summed sul- calibration AS aerosol (see Sect. 3.1). However, it is unlikely − fate peaks. Collections of less than 0.1 × 10 9 cm3 of material were that the calibration particles could become contaminated by excluded. Only experiments for which the collected mass for the ammonia to the extent that they achieve a base : acid ratio two polarities were within a factor of 2 are included in the plot, and higher than that of the stable AS salt (i.e., 2 : 1), which is the ratios are scaled to correct for differences in sampled mass. The roughly the expected composition of the calibration aerosol. filled symbols are experiments with nominally 10 ppt DMA and the So, while the size-selection process significantly impacted open symbols are with 40 ppt DMA. Blue symbols indicate periods the composition of the acidic chamber particles, this process with 1 % H2 added to the TDCIMS inlet and red symbols are with- was unlikely to influence the composition of the calibration out H2. The vertical error bars represent the standard error of the particle ion signals, and the horizontal error bars indicate the range aerosol greatly. of particle diameters that contributed to the collected mass (min- imum 2 % contribution). The TDCIMS results show consistently lower base : acid ratios than inferred by the HTDMA, but they show 5 Discussion a similar size dependence and high variability for the smallest de- tectable sizes. 5.1 Particle-phase dimethylaminium–ammonium ratios sistent bias in base : acid ratios is evident when H2 was added During these experiments, contaminant ammonia was (see next paragraph, and Fig. 10). Overall, therefore, there is present in the chamber at significant levels as a result of nothing to indicate that the measured base : acid ratios were earlier experiments with intentional ammonia additions. TD- strongly biased by sampling artifacts. CIMS measurements indicate that this ammonia was incor- Ratios of the sum of base (DMA and ammonia) to acid porated into growing sulfuric acid particles of ∼ 10–30 nm (sulfate) integrated peaks were calculated to assess the acid- with an efficiency comparable to or greater than DMA. Ob- ity of the particles in each experiment (Fig. 10). Because servations of nucleated clusters in the CLOUD chamber acid and base compounds are each detected in a different po- show that DMA is an important constituent of recently nu- larity (negative and positive, respectively), there is always cleated particles when it is present in the chamber at concen- a time gap between the observations, during which the par- trations as small as 5 ppt (Almeida et al., 2013; Kürten et al., ticle size and potentially composition have changed. How- 2014), whereas ammonia was only detected in clusters with ever, reasonably consistent results were achieved when com- more than 7 H2SO4 molecules (Bianchi et al., 2014). How- parisons were made for collections close in time for similar ever, it is possible that ammonia molecules were present in volume median diameters of sampled particles, with similar smaller clusters in the chamber and evaporated during anal- amounts of collected mass. There was not a strong relation- ysis in the APi-TOF of CI-APi-TOF instruments. The abun- ship between gas-phase sulfuric acid concentration and par- dance of ammonia in the TDCIMS-measured particles indi- ticle composition. Kim et al. (2016) similarly found a non- cates either that the molecular clusters that lead to > 10 nm

Atmos. Chem. Phys., 16, 13601–13618, 2016 www.atmos-chem-phys.net/16/13601/2016/ M. J. Lawler et al.: Acidic nanoparticles at CLOUD 13611 particles contain more ammonia than is evident from cluster sion time into an aqueous 10 nm particle is very short com- observations or that ammonia is incorporated into particles pared to the lifetime of a particle in the chamber. However, by a process distinct from cluster formation, or both. recent work on nanoparticle phase suggests that these small A particle composition model based on chemical equilib- particles should have a liquid phase (Cheng et al., 2015). Su- ria for aqueous well-mixed particles predicts roughly equal persaturated AS nanoparticles, such as those likely to be pro- + + fractions of DMAH and NH4 ions in the particles when duced in new particle formation, are expected to be liquid 15 ppt ammonia and 10 ppt DMA are present in the gas or mixed phase (depending on particle size) at the tempera- phase (Ahlm et al., 2016). At smaller DMA concentrations ture and sizes considered in these experiments (see Sect. 5 ammonium is expected to dominate. For gas-phase concen- for more details). The growth rates of 2 nm particles in the trations of 40 ppt DMA and 15 ppt ammonia, however, the CLOUD7 experiments are around 10 times faster than would particulate-phase base should be primarily DMAH+ with a be predicted if the growth were due to perfectly efficient con- + + DMAH : NH4 ratio of about 5 due to the higher basicity densation of the measured gas-phase sulfuric acid monomers and lower saturation vapor pressure of DMA (Ahlm et al., (Lehtipalo et al., 2016). This suggests that cluster–particle + + 2016). The relative DMAH : NH4 ratio is thus expected to collisions were the main driver of the growth of the newly be highly dependent on the gas-phase concentration of DMA, formed sub-2 nm particles. The calculations presented by a feature which is not observed in the TDCIMS data. This Ahlm et al. (2016) show that coagulation was also an im- may be explained in part by the fact that the DMA : NH3 ra- portant factor driving the growth of the larger particles. The tio in the chamber, as measured by IC, was actually compara- composition of the smaller clusters and particles could there- ble in both the 40 and 10 ppt DMA experiments. That is, the fore be reflected in the composition of the growing particles, ammonia contamination was somewhat lower for the 10 ppt if the timescale of the growth was significantly shorter than DMA experiments. However, for DMA : NH3 gas-phase ra- the equilibration timescale of the particles with the gas-phase + + tios above 1, there should be more DMAH than NH4 in bases (such as would occur for high-viscosity, non-aqueous the particles. The particle-phase DMAH+ : NH+ ratios are particles). However, there is no evidence indicating an en- 4 + + close to 1 in many cases, such that a relatively small differ- hancement of NH4 relative to DMAH in smaller particles ence in TDCIMS sensitivity to the two bases, or the relatively and clusters, and, according to present understanding, most large uncertainty in the gas-phase DMA : NH3 ratio, would particles should have reached thermodynamic equilibrium in be enough to remove disagreement from expectations. How- the chamber before sampling. ever, there are several cases for which the DMAH+ signal Experimental errors that would affect these interpretations is constrained by error bars to be less than two-thirds of the are limited to differential measurement sensitivity between + DMAH+ and NH+ in the particle phase and between DMA NH4 signal and, based on calibrations, the instrument sen- 4 + sitivity to DMAH appears to be the same or higher than to and NH3 in the total (mostly gas-phase) observations. In the + case of the TDCIMS particle-phase measurements, this could NH4 . This unexpected result has at least two potential expla- nations: (1) a deficiency in the thermodynamic calculations arise from a sampling artifact involving either preferential of the particle-phase composition, such as a kinetic limitation loss of DMA relative to ammonia or contamination of the for surface uptake of the bases or water; or (2) a measurement sample with ammonia. DMA is expected to form a more error or artifact in the particle- or gas-phase composition. In stable salt with sulfuric acid than ammonia, making the for- the following we discuss each of these possibilities. mer possibility unlikely. Calibrations with known standards Models that are based on chemical equilibria for well- also show no relative deficiency of the DMA signal (see mixed aqueous particles may not adequately describe Sect. 3.1). Significant contamination of the particles by am- nanoparticles. Recently formed sulfuric acid nanoparticles monia within the TDCIMS is also unlikely given the insensi- + + may have structures and corresponding intermolecular inter- tivity of the measured base : acid ratios and DMAH : NH4 actions that are different from those in a well-mixed solution. ratios to the ammonium background measured by the TD- For example, steric effects that favor the presence of ammo- CIMS. Ammonium backgrounds were over an order of mag- nia over DMA may be important at these small sizes. Ther- nitude higher when H2 gas was added, but the particle-phase modynamic calculations and previous experimental studies signals were not measurably influenced (Figs. 9 and 10). Fur- on larger particles suggest aqueous mixtures of aminium and thermore, when H2 gas was not added to the TDCIMS in- ammonium sulfates with base : acid ratios from 2 : 1 to 1 : 1 let, most of the background ammonium signal came from are the stable composition at the RH of the CLOUD chamber the chamber. After the high-temperature chamber cleaning, experiments (Chan and Chan, 2013; Sauerwein et al., 2015). which did not affect the TDCIMS (it was disconnected), the However, there were unfortunately no direct observations on ammonium background went down by a factor of 5. This in- the phase and water content of the nanoparticles in these ex- dicates that most of the ammonia that was available to be periments. picked up by particles was provided by desorption from the Kinetic limitations to surface uptake of the bases are prob- chamber walls, not by contamination within the instrument. ably only possible for non-aqueous particles, since the diffu- The gas-phase DMA observations could have been overes- timated with respect to ammonia if the DMA was preferen- www.atmos-chem-phys.net/16/13601/2016/ Atmos. Chem. Phys., 16, 13601–13618, 2016 13612 M. J. Lawler et al.: Acidic nanoparticles at CLOUD tially in the particle phase and ammonia was not, and if the and may therefore be less prominent in the APi-TOF spectra. IC system was sensitive to both particle-phase and gas-phase It should be noted that equilibrium cluster distribution mea- DMA. The TDCIMS measurements are not sensitive to gas- sured by the APi-TOF gives essentially no information about phase compounds, however, and they indicate that there was the relative importance of the different clusters for nanopar- not preferential particle uptake of DMA compared with am- ticle growth due to a lack of knowledge about the cluster for- monia. Nonetheless, for the larger mass collections, a sig- mation rates and lifetimes. In any case, it appears clear that nificant fraction of both the DMA and NH3 present in the the growth of particles above 5 nm is not due to a stepwise chamber was present in the particle phase. addition of H2SO4 and DMA molecules as appears to be the + + To summarize, the DMAH : NH4 ratios in the parti- case for the smallest clusters. cles appear to be lower than expected based on thermody- In view of the apparent production of H2SO4 in the namic equilibrium calculations. However, there is significant UPCs, it is necessary to consider the possibility that the low uncertainty in the actual gas-phase mixing ratios of DMA base : acid ratios resulted from contamination of the particles and ammonia in the chamber, and it is also possible that by sulfuric acid within the instrument. There are two dis- the TDCIMS showed differential sensitivity with respect to tinct periods when such contamination could have occurred: + + DMAH and NH4 in the salts that constituted the particles (1) while the particles were within the unipolar chargers and in the chamber. Given these uncertainties, we cannot exclude (2) while the particles rested on the precipitator wire after be- + + the possibility of no deviation in the DMAH : NH4 ratios ing collected. Significant growth in the first, brief period can from thermodynamic expectations. However, the observed be excluded by comparing the appearance times of particles base : acid ratios in the particles were far from their expected of a given size in the chamber with those for the particles values, as discussed below. exiting the UPCs, as measured with the differential mobil- ity analyzer and CPC downstream of the UPCs. There was 5.2 Particle-phase base : acid ratios no discernible growth during this period. Addition of com- pounds to the precipitator wire should be inefficient due to The TDCIMS results indicate a base : acid ratio below 0.5 for the flow of N2 sheath gas around it. Such contamination does particles 10 nm and above, despite a significant excess of gas- nonetheless occur, but its effects are in general subtracted phase DMA and ammonia with respect to sulfuric acid. This out by the background correction. A scenario in which this is counter to the expectation that the particles should reach background correction would not be sufficient would be if thermodynamic equilibrium and be closer to pH-neutral, the gas-phase H2SO4 were more efficiently incorporated into with more ammonium and dimethylaminium ions than sul- sampled particles on the wire than onto a particle-free wire. fate ions (Ahlm et al., 2016). This discrepancy may be ex- Figure 3 suggests that this was not a significant issue, since plained by uncertainties in the thermodynamics of small par- there was no evidence for increased sulfate signal in chamber ticles, but oxidative chemistry may also play a role in this samples with respect to calibration standards of similar sam- case. Possible experimental biases are discussed below as pled mass. The estimation of collected particle mass cannot well. be influenced by processes occurring after collection onto the As mentioned previously, CLOUD measurements and nu- wire. merical calculations of cluster and nanoparticle growth show The TDCIMS data overall indicate a large increase of base that cluster–cluster, cluster–particle, and particle–particle fraction as the particles grew from 10 to 20 nm (Fig. 10), collisions are important drivers of growth in the DMA– though variability among individual experiments was high sulfuric acid system (Lehtipalo et al., 2016; Ahlm et al., for ∼ 10 nm particle composition. The direction of this com- 2016). The composition of clusters could therefore be im- position change over this range of diameter is corroborated portant for determining the composition of > 10 nm particles by aerosol hygroscopicity results from the same experiments. for timescales shorter than the equilibration timescales of These measurements showed that 15 nm particles had sig- the particles. Positive ion APi-TOF mass spectrometer mea- nificantly lower hygroscopicity than 10 nm particles. Kim surements indicate that up to the maximum observable sizes et al. (2016) estimated a base : acid ratio of roughly 1.0 for for this technique (∼ 2 nm), the composition of ammonia– 15 nm particles and 0.3 for 10 nm particles, based on the DMA–sulfuric acid clusters may be acidic for the small- measured hygroscopicities. Our results indicate even lower est clusters but approaches a 1.2 : 1 base : acid composition ratios. Given the uncertainty of TDCIMS sensitivities to very (Bianchi et al., 2014; Fig. 10). It seems possible that the least- acidic particles, the TDCIMS ion ratio results are estimates neutralized (most acidic) clusters could have the highest ac- below a base : acid ratio of ∼ 1.0. The differences in the ab- commodation coefficients and therefore be the most impor- solute base : acid ratio between the TDCIMS and HTDMA- tant clusters for nanoparticle growth, but we could find no ev- based measurements could be related to the handling of the idence to support this in the literature. This hypothesis could particles during sampling. In the case of the TDCIMS ob- explain why the observed > 5 nm particles are more acidic servations, the sampled particles were exposed to clean N2 than most of the observed < 2 nm clusters. The most effec- gas for up to 30 min. This is unlikely to introduce much con- tive clusters may have shorter lifetimes to accommodation tamination but may result in the loss of molecules that do

Atmos. Chem. Phys., 16, 13601–13618, 2016 www.atmos-chem-phys.net/16/13601/2016/ M. J. Lawler et al.: Acidic nanoparticles at CLOUD 13613 not form a stable salt while the particles dry in the N2 flow. ence between 10 nm and larger particles, through a few possi- The final composition and phase of the drying particles is ble mechanisms. There may be an enhancement of interfacial uncertain and probably depends on size. However, the many and/or aqueous-phase reaction rates for the smaller particles calibrations performed during and after CLOUD7 indicate relative to the larger, phase-separated particles. In addition, that the TDCIMS analysis does not show a strong tendency base molecules present in the aqueous phase are more likely to lose bases prior to analysis for ammonium or dimethy- to be lost as the particles evaporate during sampling than base laminium sulfate aerosol (see Sect. 3.1). That the base : acid molecules held in salts in a solid phase, which must conform ratio is not strongly biased within the instrument is shown by to a regular repeating chemical structure. Therefore, larger two observations of very different particle-phase base : acid phase-separated particles may be better able to maintain their ratios from most of the DMA–H2SO4 experiments described base molecules when sampled by the TDCIMS. However, the here: (a) we measured calibration aerosol with much higher HTDMA measurements also show a lower base : acid ratio in base : acid ratios than the aerosol formed in the experiments the 10 nm particles, and sampled particles in this instrument described here (see Sect. 3.1); (b) when we exposed the par- do not experience a long period in dry, clean air during which ticles to some air of unknown composition by size-selecting base molecules can slowly evaporate. Overall, the evidence them in one experiment, we contaminated the particles with suggests that the apparent high acidity of the small particles ammonium and found the base : acid ratio to increase (see is a feature of the chamber particles and not an artifact of the Sect. 4). The operation of the HTDMA instrument depends experimental methods. on the addition of humidified and dried HEPA-filtered air. It We can also consider whether the high chamber SO2 mix- seems likely that these observations of base : acid ratio could ing ratios could have led to low base : acid ratios in the par- be biased higher than the TDCIMS observations due to con- ticles by rapid sulfate production which outstripped the abil- tamination by ammonia. In any case, the independent ob- ity of the gas-phase bases to neutralize the nanoparticles. servations of particle chemistry and physical properties pro- Ahlm et al. (2016) found that for sub-10 nm particles, ob- vided by the TDCIMS and HTDMA agree that particles at served growth rates were a factor of 2 to 5 times as large as 10–20 nm sizes had a substantially larger base ion fraction predicted when considering growth by H2SO4 condensation, than particles near 10 nm. cluster collisions, and coagulation with other nanoparticles. The large change in particle composition around 10 nm This is consistent with the possibility that additional chem- indicates a sharp change in particle compositional growth. istry played a role in nanoparticle growth in this system. SO2, This could result from a phase change from a purely liq- or S(IV), can be oxidized to sulfate in aqueous droplets by uid state to a mixed-phase state with a solid core. Recent H2O2 and O3 (e.g., Caffrey et al., 2001; Hoyle et al., 2016). results have shown that the phase state of nanoparticles is We are aware of only one study examining the potential for size dependent, such that particles are expected to be com- known aqueous-phase S(IV) oxidation reactions to cause sig- pletely liquid below some size threshold, even for supersatu- nificant particle growth in nanoparticles (Kerminen et al., rated droplets (Cheng et al., 2015). The location of this tran- 2000). This numerical study showed that known pathways sition depends on particle size, composition, water content, cannot result in enough growth for newly formed nanopar- and temperature, and it was demonstrated to occur for AS ticles to reach CCN size, given typical tropospheric particle nanoparticles at around 10 nm at 298 K for a particulate AS lifetimes and oxidant and SO2 mixing ratios of 100 ppt. How- mass fraction of 0.63 (Cheng et al., 2015). At the slightly ever, the SO2 mixing ratio in the present experiments was lower temperature in the present study (273 K), this transition 63 ppb, typical of pollution plumes, and there was no larger occurs at a slightly smaller particle diameter, closer to 9 nm. aerosol mode to act as a coagulation sink for the small parti- For the dimethylammonium–ammonium–sulfate nanoparti- cles and thereby limit their growth. Furthermore, H2O2 lev- cles studied here, a similar size-dependent transition seems els were also probably higher than typical ambient levels, as plausible. However, less is known about the thermodynam- this species is the main HO2 radical sink under the CLOUD ics of these mixed ammonium–aminium sulfate salts. Even chamber conditions of extremely low NOx. For particles of for large droplets, where size-dependent effects are not im- pH < 4 and for low liquid water content, as in the case of the portant for phase state, pure dimethylaminium sulfate does recently formed particles in this study, H2O2 is the dominant not form a solid phase, even at < 3 % RH. However, the ex- oxidant of S(IV) in the atmosphere. To estimate the potential posure of such an amorphous liquid droplet to ammonia re- of aqueous-phase S(IV) oxidation to contribute to nanoparti- sults in the formation of a solid phase which contains am- cle growth, we considered the rate of H2O2 + S(IV) reaction monium and aminium and is resistant to further exchange in a 5.4 nm diameter ammonium bisulfate particle as a proxy (Chan and Chan, 2013). If a phase transition does occur for for the chamber particles. We used the Extended Aerosol the mixed ammonium–dimethylaminium–sulfate nanoparti- Thermodynamics Model (E-AIM) to estimate the pH and as- cles in the present study, the molar concentrations of the sumed that [SO2] and [H2O2] were in Henry’s law equilib- species present in the smaller, purely liquid particles would rium with the gas phase. For a gas-phase H2O2 level of 1 ppb, be higher than those in the liquid phase of the slightly larger S(IV) oxidation was estimated at about 0.002 molecules s−1 mixed-phase particles. This could explain the strong differ- for the small particle. For comparison, H2SO4 collisions with www.atmos-chem-phys.net/16/13601/2016/ Atmos. Chem. Phys., 16, 13601–13618, 2016 13614 M. J. Lawler et al.: Acidic nanoparticles at CLOUD

−1 the particle would occur at a rate of about 0.2 molecules s concentrations but SO2 levels remain high. This is consistent 7 −3 for an H2SO4 concentration of 1 × 10 cm , with still larger with the findings of Xue et al. (2016), who show that a di- contributions from cluster collisions and coagulation. This urnally unvarying SO2–surface reaction is likely an impor- aqueous chemistry was therefore an insignificant contributor tant contributor to secondary aerosol formation in haze–fog to the particle growth and composition. events in Chinese megacities and that it represents a signifi- However, in addition to the well-known aqueous-phase cant fraction (about a third) of the sulfate production at night. S(IV) oxidation, a recent study has demonstrated an ex- SO2 levels for these conditions were similar to the present tremely rapid surface reaction of SO2 on acidic micro- study, in the range of tens of ppb. For comparison, if we as- droplets to produce sulfate and other oxidized sulfur species sume a 50 ppb SO2 mixing ratio and a particle distribution for in the condensed phase, in the absence of any added oxi- which the available particle-phase surface area is represented dants other than O2 (Hung and Hoffman, 2015). The ob- by 100 nm diameter condensation nuclei with a concentra- −3 served process was fastest for droplets with a pH ∼ 3, for tion of 5000 cm , a 0.05 % SO2–surface reaction probabil- which the reaction rate was about 4 orders of magnitude ity leads to a 5 µg m−3 sulfate production rate, which is of faster than S(IV) oxidation for typical ambient H2O2 levels. the right order of magnitude to explain the polluted megacity Several recent studies indicate a likely role for such hetero- observations. However, this surface mechanism is much less geneous SO2 conversion to sulfate in major haze–fog events likely to be important for new particle formation in remote, in polluted Chinese megacities (e.g., Xue et al., 2016, and unpolluted areas. For example, at the well-studied Hyytiälä references therein). Since the SO2 mixing ratio was over 3 site in the Finnish boreal forest, SO2 concentrations around orders of magnitude larger than the DMA mixing ratio, the 0.1 ppb are typical. At this low level, the SO2–surface reac- −1 rates of nanoparticle collisions with SO2 were more frequent tion could only contribute about 0.1 nm h to growth, which than collisions with DMA of NH3 by a similar or even much is a small fraction of typical growth rates observed at this site greater factor, when considering that a significant fraction of (Yli-Juuti et al., 2011). the nitrogen bases present in the chamber were likely bound in the particle phase. Sulfur dioxide, by contrast, should be > 99.99 % in the gas phase given the high mixing ratio 6 Conclusions added to the chamber, sulfur dioxide’s relatively low solu- bility, and the low aerosol loading in these experiments. If Nanoparticles were formed by reactions involving sulfuric 0.05 % of SO2 collisions with a 5 nm particle irreversibly acid, DMA, water vapor, and contaminant ammonia. DMA formed particle-phase sulfate, even the largest discrepancies is the more effective stabilizing base for sulfuric acid during between predicted and observed growth rates could be recon- the initial nucleation and growth to at least 2 nm, but at sizes ciled. This hypothetical rate would be about 3 times as fast as around 10 nm and greater, ammonia was taken up into the nitrogen base collisions with the same particle if there were particles with comparable or even greater efficiency. Given 10 ppt of the nitrogen base (ammonia or DMA) in the gas the greater abundance of ammonia than amines in most atmo- phase, which is within the range of possibility. Under these spheric regions, ammonia may be more important than DMA conditions, the particle-phase sulfate production rate could for forming particulate sulfate salts in growing nanoparticles. exceed the rate at which nitrogen bases would be able to neu- The sulfuric acid particles were not observed to be fully neu- tralize the particles. Therefore we consider it likely that the tralized by bases, despite the presence in the gas phase of unexpectedly low base : acid ratios are explained by a rapid at least an order of magnitude more DMA and NH3 than surface reaction of SO2. H2SO4. For many particle collections at the smallest measur- able particle sizes (around 10 nm), the base : acid ratio was a 5.3 Implications for ambient nanoparticles few times lower than for ∼ 20 nm particles. This could be the result of a phase transition, in which particles moved Newly formed nanoparticles in the planetary boundary layer from a supersaturated liquid to a phase-separated particle are unlikely to contain only sulfuric acid and nitrogen bases with a solid core as they grow. The base ions (ammonium once they reach the sizes considered in this study. However, and dimethylaminium) may have been maintained more sta- by observing this chemically simple system, it was possi- bly in the solid phase of the larger particles upon sampling, ble to probe what appears to be a rapid heterogeneous re- and we also consider it possible that the chemistry of parti- action which may play a role in some atmospheric regions. cle growth was different for the particles of different phase If the inferred reaction probability for SO2 is applicable in states due to the different ion concentrations in their mo- other circumstances, then this reaction is competitive with bile phases. However, further work is needed to character- H2SO4 uptake when the SO2 concentration is about a fac- ize the phase states of mixed ammonium–aminium–sulfate tor of 2000 times larger than the H2SO4 concentration (e.g., nanoparticles to assess these possibilities and to understand 7 −3 about 1 ppb SO2 and 1 × 10 cm at sea level). This is most the growth pathways of such particles. The very low partic- likely to occur at night in regions with strong SO2 emissions, ulate base : acid ratios observed do not have an unequivocal when short-lived, photochemically produced H2SO4 is at low explanation, but we propose that the sampled particles were

Atmos. Chem. Phys., 16, 13601–13618, 2016 www.atmos-chem-phys.net/16/13601/2016/ M. J. Lawler et al.: Acidic nanoparticles at CLOUD 13615 not at thermodynamic equilibrium with the gas-phase vapors Heinritzi, M., Henschel, H., Jokinen, T., Junninen, H., Kajos, M., due to a rapid SO2–surface reaction. These observations sug- Kangasluoma, J., Keskinen, H., Kupc, A., Kurtén, T., Kvashin, gest that effective stabilization of sulfuric acid in nanoparti- A. N., Laaksonen, A., Lehtipalo, K., Leiminger, M., Leppä, cles requires fewer base molecules than would be in a 2 : 1 J., Loukonen, V., Makhmutov, V., Mathot, S., McGrath, M. J., or even a 1 : 1 salt. The very low base : acid ratios observed Nieminen, T., Olenius, T., Onnela, A., Petäjä, T., Riccobono, F., in these laboratory-generated particles may not be expected Riipinen, I., Rissanen, M., Rondo, L., Ruuskanen, T., Santos, F. D., Sarnela, N., Schallhart, S., Schnitzhofer, R., Seinfeld, J. H., during new particle formation under more typical ambient Simon, M., Sipilä, M., Stozhkov, Y., Stratmann, F., Tomé, A., conditions of lower SO2 and higher NOx. Tröstl, J., Tsagkogeorgas, G., Vaattovaara, P., Viisanen, Y., Vir- tanen, A., Vrtala, A., Wagner, P. E., Weingartner, E., Wex, H., Williamson, C., Wimmer, D., Ye, P., Yli-Juuti, T., Carslaw, K. 7 Data availability S., Kulmala, M., Curtius, J., Baltensperger, U., Worsnop, D. R., Vehkamäki, H., and Kirkby, J.: Molecular understanding of sul- Particle composition and size distribution data are available phuric acid-amine particle nucleation in the atmosphere, Nature, through contact with the authors ([email protected]). 502, 359–363, doi:10.1038/nature12663, 2013. Barsanti, K. C., McMurry, P. H., and Smith, J. N.: The potential con- tribution of organic salts to new particle growth, Atmos. Chem. Acknowledgements. We would like to thank CERN for support- Phys., 9, 2949–2957, doi:10.5194/acp-9-2949-2009, 2009. ing CLOUD with important technical and financial resources, Bianchi, F., Praplan, A. P., Sarnela, N., Dommen, J., Ku, A., Ortega, and for providing a particle beam from the CERN Proton Syn- I. K., Schobesberger, S., Junninen, H., Simon, M., Tro, J., Joki- chrotron. This research has received funding from the EC Seventh nen, T., Sipila, M., Adamov, A., Amorim, A., Almeida, J., Breit- Framework Programme (Marie Curie Initial Training Network enlechner, M., Duplissy, J., Ehrhart, S., Flagan, R. C., Franchin, “CLOUD-ITN” grant no. 215072, the ERC-Advanced grant A., Hakala, J., Hansel, A., Heinritzi, M., Kangasluoma, J., Kesk- “ATMNUCLE” (no. 227463), the German Federal Ministry of inen, H., Kim, J., Kirkby, J., Laaksonen, A., Lawler, M. J., Lehti- Education and Research (project no. 01LK0902A), the Swiss palo, K., Leiminger, M., Makhmutov, V., Mathot, S., Onnela, A., National Science Foundation (project nos. 206621 125025 and Peta, T., Riccobono, F., Rissanen, M. P., Rondo, L., Tome, A., 206620 130527), the Academy of Finland Centre of Excellence Virtanen, A., Viisanen, Y., Williamson, C., Wimmer, D., Win- program (project no. 1118615), Academy of Finland (project kler, P. M., Ye, P., Curtius, J., Kulmala, M., Worsnop, D. R., no. 138951), the Austrian Science Fund (FWF; project no. J3198- Donahue, N. M., and Baltensperger, U.: Insight into Acid–Base N21), the Portuguese Foundation for Science and Technology Nucleation Experiments by Comparison of the Chemical Com- (project no. CERN/FP/116387/2010), the US National Science position of Positive, Negative, and Neutral Clusters, Environ. Sci. Foundation, and the Russian Foundation for Basic Research (grant Technol., 48, 13675–13684, doi:10.1021/es502380b, 2014. N08-02-91006-CERN). James N. Smith acknowledges funding Bzdek, B. R., Ridge, D. P., and Johnston, M. V.: Amine exchange from the Finnish Academy (Grant No. 251007) and US Department into ammonium bisulfate and ammonium nitrate nuclei, Atmos. of Energy (grant no. DE-SC0014469). The National Center for Chem. Phys., 10, 3495–3503, doi:10.5194/acp-10-3495-2010, Atmospheric Research is sponsored by the National Science 2010. Foundation. Bzdek, B. R., Zordan, C. A., Pennington, M. R., Luther, G. W., and Johnston, M. V.: Quantitative assessment of the sulfuric acid Edited by: J. Liggio contribution to new particle growth., Environ. Sci. Technol., 46, Reviewed by: two anonymous referees 4365–4373, doi:10.1021/es204556c, 2012. Bzdek, B. R., Lawler, M. J., Horan, A. J., Pennington, M. R., Depalma, J. W., Zhao, J., Smith, J. N., and John- ston, M. V.: Molecular Constraints on Particle Growth during References New Particle Formation, Geophys. Res. Lett., 41, 6045–6054, doi:10.1002/2014GL060160, 2014. Ahlm, L., Yli-Juuti, T., Schobesberger, S., Praplan, A. P., Kim, J., Caffrey, P., Hoppel, W., Frick, G., Pasternack, L., Fitzgerald, J., Tikkanen, O.-P., Lawler, M. J., Smith, J. N., Tröstl, J., Acosta Hegg, D., Gao, S., Leaitch, R., Shantz, N., Albrechcinski, T., Navarro, J. C., Baltensperger, U., Bianchi, F., Donahue, N. M., and Ambrusko, J.: In-cloud oxidation of SO2 by O3 and H2O2: Duplissy, J., Franchin, A., Jokinen, T., Keskinen, H., Kürten, A., cloud chamber measurements and modeling of particle growth, J. Laaksonen, A., Lehtipalo, K., Petäjä, T., Riccobono, F., Rissa- Geophys. Res., 106, 27587–27601, doi:10.1029/2000JD900844, nen, M. P., Rondo, L., Schallhart, S., Simon, M., Winkler, P. M., 2001. Worsnop, D. R., Virtanen, A., and Riipinen, I.: Modeling the Chan, L. P. and Chan, C. K.: Role of the Aerosol Phase State in thermodynamics and kinetics of sulfuric acid-dimethylamine- Ammonia/Amines Exchange Reactions, Environ. Sci. Technol., water nanoparticle growth in the CLOUD chamber, Aerosol Sci. 47, 5755–5762, doi:10.1021/es4004685, 2013. Tech., doi:10.1080/02786826.2016.1223268, online first, 2016. Chen, D.-R. and Pui, D.: A high efficiency, high throughput unipolar Almeida, J., Schobesberger, S., Kürten, A., Ortega, I. K., aerosol charger for nanoparticles, J. Nanopart. Res., 1, 115–126, Kupiainen-Määttä, O., Praplan, A. P., Adamov, A., Amorim, A., 1999. Bianchi, F., Breitenlechner, M., David, A., Dommen, J., Don- ahue, N. M., Downard, A., Dunne, E., Duplissy, J., Ehrhart, S., Flagan, R. C., Franchin, A., Guida, R., Hakala, J., Hansel, A., www.atmos-chem-phys.net/16/13601/2016/ Atmos. Chem. Phys., 16, 13601–13618, 2016 13616 M. J. Lawler et al.: Acidic nanoparticles at CLOUD

Cheng, Y., Su, H., Koop, T., Mikhailov, E., and Pöschl, U.: Size Atmos. Chem. Phys., 16, 1693–1712, doi:10.5194/acp-16-1693- dependence of phase transitions in aerosol nanoparticles, Nat. 2016, 2016. Commun., 6, 5923, doi:10.1038/ncomms6923, 2015. Hung, H. and Hoffmann, M.: Oxidation of Gas-Phase SO2 Duplissy, J., Merikanto, J., Franchin, A., Tsagkogeorgas, G., Kan- on the Surfaces of Acidic Micro-Droplets: Implications for gasluoma, J., Wimmer, D., Vuollekoski, H., Schobesberger, Sulfate and Sulfate Radical Anion Formation in the Atmo- S., Lehtipalo, K., Flagan, R. C., Brus, D., Donahue, N. M., spheric Liquid Phase, Environ. Sci. Technol., 49, 13768–13776, Vehkämäki, H., Almeida, J., Amorim, A., Barmet, P., Bianchi, doi:10.1021/acs.est.5b01658, 2015. F., Breitenlechner, M., Dunne, E. M., Guida, R., Henschel, H., Jokinen, T., Sipilä, M., Junninen, H., Ehn, M., Lönn, G., Hakala, Junninen, H., Kirkby, J., Kürten, A., Kupc, A., Määttänen, A., J., Petäjä, T., Mauldin III, R. L., Kulmala, M., and Worsnop, Makhmutov, V., Mathot, S., Nieminen, T., Onnela, A., Pra- D. R.: Atmospheric sulphuric acid and neutral cluster measure- plan, A. P., Riccobono, F., Rondo, L., Steiner, G., Tome, A., ments using CI-APi-TOF, Atmos. Chem. Phys., 12, 4117–4125, Walther, H., Baltensperger, U., Carslaw, K. S., Dommen, J., doi:10.5194/acp-12-4117-2012, 2012. Hansel, A., Petäjä, T., Sipilä, M., Stratmann, F., Vrtala, A., Wag- Kazil, J., Stier, P., Zhang, K., Quaas, J., Kinne, S., O’Donnell, D., ner, P. E., Worsnop, D. R., Curtius, J., and Kulmala, M.: Ef- Rast, S., Esch, M., Ferrachat, S., Lohmann, U., and Feichter, fect of ions on sulfuric acid-water binary particle formation II: J.: Aerosol nucleation and its role for clouds and Earth’s ra- Experimental data and comparison with QC-normalized classi- diative forcing in the aerosol-climate model ECHAM5-HAM, cal nucleation theory, J. Geophys. Res.-Atmos., 121, 1752–1775, Atmos. Chem. Phys., 10, 10733–10752, doi:10.5194/acp-10- doi:10.1002/2015JD023539, 2016. 10733-2010, 2010. Ehn, M., Thornton, J. A., Kleist, E., Sipilä, M., Junninen, H., Pulli- Kerminen, V., Pirjola, L., Boy, M., Eskola, A., Tenilä, K., Laakso, nen, I., Springer, M., Rubach, F., Tillmann, R., Lee, B., Lopez- L., Asmi, A., Hienola, J., Lauri, A., Vainio, V., Lehtinen, K., Hilfiker, F., Andres, S., Acir, I.-H., Rissanen, M., Jokinen, T., and Kulmala, M.: Interaction between SO2 and submicron at- Schobesberger, S., Kangasluoma, J., Kontkanen, J., Nieminen, mospheric particles, Atmos. Res., 54, 41–57, 2000. T., Kurtén, T., Nielsen, L. B., Jørgensen, S., Kjaergaard, H. G., Keskinen, H., Virtanen, A., Joutsensaari, J., Tsagkogeorgas, G., Du- Canagaratna, M., Maso, M. D., Berndt, T., Petäjä, T., Wahner, plissy, J., Schobesberger, S., Gysel, M., Riccobono, F., Slowik, J. A., Kerminen, V.-M., Kulmala, M., Worsnop, D. R., Wildt, J., G., Bianchi, F., Yli-Juuti, T., Lehtipalo, K., Rondo, L., Breiten- and Mentel, T. F.: A large source of low-volatility secondary or- lechner, M., Kupc, A., Almeida, J., Amorim, A., Dunne, E. M., ganic aerosol, Nature, 506, 476–479, doi:10.1038/nature13032, Downard, A. J., Ehrhart, S., Franchin, A., Kajos, M. K., Kirkby, 2014. J., Kürten, A., Nieminen, T., Makhmutov, V., Mathot, S., Mietti- Franchin, A., Ehrhart, S., Leppä, J., Nieminen, T., Gagné, S., nen, P., Onnela, A., Petäjä, T., Praplan, A., Santos, F. D., Schall- Schobesberger, S., Wimmer, D., Duplissy, J., Riccobono, F., hart, S., Sipilä, M., Stozhkov, Y., Tomé, A., Vaattovaara, P., Wim- Dunne, E. M., Rondo, L., Downard, A., Bianchi, F., Kupc, A., mer, D., Prevot, A., Dommen, J., Donahue, N. M., Flagan, R. C., Tsagkogeorgas, G., Lehtipalo, K., Manninen, H. E., Almeida, J., Weingartner, E., Viisanen, Y., Riipinen, I., Hansel, A., Curtius, Amorim, A., Wagner, P. E., Hansel, A., Kirkby, J., Kürten, A., J., Kulmala, M., Worsnop, D. R., Baltensperger, U., Wex, H., Donahue, N. M., Makhmutov, V.,Mathot, S., Metzger, A., Petäjä, Stratmann, F., and Laaksonen, A.: Evolution of particle compo- T., Schnitzhofer, R., Sipilä, M., Stozhkov, Y., Tomé, A., Kermi- sition in CLOUD nucleation experiments, Atmos. Chem. Phys., nen, V.-M., Carslaw, K., Curtius, J., Baltensperger, U., and Kul- 13, 5587–5600, doi:10.5194/acp-13-5587-2013, 2013. mala, M.: Experimental investigation of ion–ion recombination Kim, J., Ahlm, L., Yli-Juuti, T., Lawler, M., Keskinen, H., Tröstl, under atmospheric conditions, Atmos. Chem. Phys., 15, 7203– J., Schobesberger, S., Duplissy, J., Amorim, A., Bianchi, F., Don- 7216, doi:10.5194/acp-15-7203-2015, 2015. ahue, N. M., Flagan, R. C., Hakala, J., Heinritzi, M., Jokinen, T., Glasoe, W. A., Volz, K., Panta, B., Freshour, N., Bachman, Kürten, A., Laaksonen, A., Lehtipalo, K., Miettinen, P., Petäjä, R., Hanson, D. R., Mcmurry, P. H., and Jen, C.: Sulfu- T., Rissanen, M. P., Rondo, L., Sengupta, K., Simon, M., Tomé, ric acid nucleation: An experimental study of the effect A., Williamson, C., Wimmer, D., Winkler, P. M., Ehrhart, S., Ye, of seven bases, J. Geophys. Res.-Atmos., 120, 1933–1950, P., Kirkby, J., Curtius, J., Baltensperger, U., Kulmala, M., Lehti- doi:10.1002/2014JD022730, 2015. nen, K. E. J., Smith, J. N., Riipinen, I., and Virtanen, A.: Hy- He, F. and Hopke, P. K.: SO2 Oxidation and H2O-H2SO4 Binary groscopicity of nanoparticles produced from homogeneous nu- Nucleation by Radon Decay, Aerosol Sci. Tech., 23, 411–421, cleation in the CLOUD experiments, Atmos. Chem. Phys., 16, doi:10.1080/02786829508965324, 1995. 293–304, doi:10.5194/acp-16-293-2016, 2016. Hoyle, C. R., Fuchs, C., Järvinen, E., Saathoff, H., Dias, A., El Kirkby, J., Curtius, J., Almeida, J., and Dunne, E.: Role of sulphuric Haddad, I., Gysel, M., Coburn, S. C., Tröstl, J., Bernhammer, acid, ammonia and galactic cosmic rays in atmospheric aerosol A.-K., Bianchi, F., Breitenlechner, M., Corbin, J. C., Craven, J., nucleation, Nature, 476, 429–433, doi:10.1038/nature10343, Donahue, N. M., Duplissy, J., Ehrhart, S., Frege, C., Gordon, H., 2011. Höppel, N., Heinritzi, M., Kristensen, T. B., Molteni, U., Nich- Kuang, C., McMurry, P. H., McCormick, A. V., and Eisele, F. L.: man, L., Pinterich, T., Prévôt, A. S. H., Simon, M., Slowik, J. Dependence of nucleation rates on sulfuric acid vapor concen- G., Steiner, G., Tomé, A., Vogel, A. L., Volkamer, R., Wagner, tration in diverse atmospheric locations, J. Geophys. Res., 113, A. C., Wagner, R., Wexler, A. S., Williamson, C., Winkler, P. M., 1–9, doi:10.1029/2007JD009253, 2008. Yan, C., Amorim, A., Dommen, J., Curtius, J., Gallagher, M. W., Kuang, C., Riipinen, I., Sihto, S.-L., Kulmala, M., McCormick, A. Flagan, R. C., Hansel, A., Kirkby, J., Kulmala, M., Möhler, O., V., and McMurry, P. H.: An improved criterion for new particle Stratmann, F., Worsnop, D. R., and Baltensperger, U.: Aqueous formation in diverse atmospheric environments, Atmos. Chem. phase oxidation of sulphur dioxide by ozone in cloud droplets, Phys., 10, 8469–8480, doi:10.5194/acp-10-8469-2010, 2010.

Atmos. Chem. Phys., 16, 13601–13618, 2016 www.atmos-chem-phys.net/16/13601/2016/ M. J. Lawler et al.: Acidic nanoparticles at CLOUD 13617

Kulmala, M., Riipinen, I., Sipilä, M., Manninen, H. E., Petäjä, Praplan, A. P., Bianchi, F., Dommen, J., and Baltensperger, U.: T., Junninen, H., Maso, M. D., Mordas, G., Mirme, A., Vana, Dimethylamine and ammonia measurements with ion chro- M., Hirsikko, A., Laakso, L., Harrison, R. M., Hanson, I., Le- matography during the CLOUD4 campaign, Atmos. Meas. ung, C., Lehtinen, K. E. J., and Kerminen, V.-M.: Toward direct Tech., 5, 2161–2167, doi:10.5194/amt-5-2161-2012, 2012. measurement of atmospheric nucleation., Science, 318, 89–92, Qiu, C. and Zhang, R.: Physiochemical properties of alkylaminium doi:10.1126/science.1144124, 2007. sulfates: hygroscopicity, thermostability, and density, Environ. Kupc, A., Amorim, A., Curtius, J., Danielczok, A., Duplissy, J., Sci. Technol., 46, 4474–4480, doi:10.1021/es3004377, 2012. Ehrhart, S., Walther, H., Ickes, L., Kirkby, J., Kürten, A., Lima, Riccobono, F., Schobesberger, S., Scott, C. E., Dommen, J., Ortega, J. M., Mathot, S., Minginette, P., Onnela, A., Rondo, L., and I. K., Rondo, L., Almeida, J., Amorim, A., Bianchi, F., Breiten- Wagner, P. E.: A fibre-optic UV system for H2SO4 production lechner, M., David, A., Downard, A., Dunne, E. M., Duplissy, in aerosol chambers causing minimal thermal effects, J. Aerosol J., Ehrhart, S., Flagan, R. C., Franchin, A., Hansel, A., Junninen, Sci., 42, 532–543, doi:10.1016/j.jaerosci.2011.05.001, 2011. H., Kajos, M., Keskinen, H., Kupc, A., Kürten, A., Kvashin, A. Kürten, A., Jokinen, T., Simon, M., Sipilä, M., Sarnela, N., Jun- N., Laaksonen, A., Lehtipalo, K., Makhmutov, V., Mathot, S., ninen, H., Adamov, A., Almeida, J., Amorim, A., Bianchi, F., Nieminen, T., Onnela, A., Petäjä, T., Praplan, A. P., Santos, F. Breitenlechner, M., Dommen, J., Donahue, N. M., Duplissy, J., D., Schallhart, S., Seinfeld, J. H., Sipilä, M., Spracklen, D. V, Ehrhart, S., Flagan, R. C., Franchin, A., Hakala, J., Hansel, A., Stozhkov, Y., Stratmann, F., Tomé, A., Tsagkogeorgas, G., Vaat- Heinritzi, M., Hutterli, M., Kangasluoma, J., Kirkby, J., Laakso- tovaara, P., Viisanen, Y., Vrtala, A., Wagner, P. E., Donahue, N. nen, A., Lehtipalo, K., Leiminger, M., Makhmutov, V., Mathot, M., Kirkby, J., Kulmala, M., Worsnop, D. R., and Baltensperger, S., Onnela, A., Petäjä, T., Praplan, A. P., Riccobono, F., Rissa- U.: Oxidation Products of Biogenic Emissions Contribute to Nu- nen, M. P., Rondo, L., Schobesberger, S., Seinfeld, J. H., Steiner, cleation of Atmospheric Particles, Science, 344, 717–722, 2014. G., Tomé, A., Tröstl, J., Winkler, P. M., Williamson, C., Wim- Riipinen, I., Yli-Juuti, T., Pierce, J. R., Petäjä, T., Worsnop, D. R., mer, D., Ye, P., Baltensperger, U., Carslaw, K. S., Kulmala, M., Kulmala, M., and Donahue, N. M.: The contribution of organics Worsnop, D. R., and Curtius, J.: Neutral molecular cluster forma- to atmospheric nanoparticle growth, Nat. Geosci., 5, 453–458, tion of sulfuric acid-dimethylamine observed in real time under doi:10.1038/ngeo1499, 2012. atmospheric conditions, P. Natl. Acad. Sci. USA, 111, 15019– Sauerwein, M., Clegg, S. L., and Chan, C. K.: Water Ac- 15024, doi:10.1073/pnas.1404853111, 2014. tivities and Osmotic Coefficients of Aqueous Solutions Lawler, M. J., Whitehead, J., O’Dowd, C., Monahan, C., Mc- of Five Alkyl-Aminium Sulfates and Their Mixtures ◦ Figgans, G., and Smith, J. N.: Composition of 15–85 nm par- with H2SO4 at 25 C, Aerosol Sci. Tech., 49, 566–579, ticles in marine air, Atmos. Chem. Phys., 14, 11557–11569, doi:10.1080/02786826.2015.1043045, 2015. doi:10.5194/acp-14-11557-2014, 2014. Schnitzhofer, R., Metzger, A., Breitenlechner, M., Jud, W., Hein- Lehtipalo, K., Rondo, L., Kontkanen, J., Schobesberger, S., Jokinen, ritzi, M., De Menezes, L.-P., Duplissy, J., Guida, R., Haider, S., T., Sarnela, N., Kürten, A., Ehrhart, S., Franchin, A., Nieminen, Kirkby, J., Mathot, S., Minginette, P., Onnela, A., Walther, H., T., Riccobono, F., Sipilä, M., Yli-Juuti, T., Duplissy, J., Adamov, Wasem, A., Hansel, A., and the CLOUD Team: Characterisation A., Ahlm, L., Almeida, J., Amorim, A., Bianchi, F., Breitenlech- of organic contaminants in the CLOUD chamber at CERN, At- ner, M., Dommen, J., Downard, A. J., Dunne, E. M., Flagan, R. mos. Meas. Tech., 7, 2159–2168, doi:10.5194/amt-7-2159-2014, C., Guida, R., Hakala, J., Hansel, A., Jud, W., Kangasluoma, J., 2014. Kerminen, V.-M., Keskinen, H., Kim, J., Kirkby, J., Kupc, A., Schobesberger, S., Junninen, H., Bianchi, F., Lönn, G., Ehn, M., Kupiainen-Määttä, O., Laaksonen, A., Lawler, M. J., Leiminger, Lehtipalo, K., Dommen, J., Ehrhart, S., Ortega, I. K., Franchin, M., Mathot, S., Olenius, T., Ortega, I. K., Onnela, A., Petäjä, T., A., Nieminen, T., Riccobono, F., Hutterli, M., Duplissy, J., Praplan, A. P., Rissanen, M. P., Ruuskanen, T. M., Santos, F. D., Almeida, J., Amorim, A., Breitenlechner, M., Downard, A. J., Schallhart, S., Schnitzhofer, R., Simon, M., Smith, J. N., Tröstl, Dunne, E. M., Flagan, R. C., Kajos, M., Keskinen, H., Kirkby, J., Tsagkogeorgas, G., Tomé, A., Vaattovaara, P., Vehkämäki, J., Kupc, A., Kürten, A., Kurtén, T., Laaksonen, A., Mathot, H., Vrtala, A. E., Wagner, P. E., Williamson, C., Wimmer, D., S., Onnela, A., Praplan, A. P., Rondo, L., Santos, F. D., Schall- Winkler, P. M., Virtanen, A., Donahue, N. M., Carslaw, K. S., hart, S., Schnitzhofer, R., Sipilä, M., Tomé, A., Tsagkogeorgas, Baltensperger, U., Riipinen, I., Curtius, J., Worsnop, D. R., and G., Vehkamäki, H., Wimmer, D., Baltensperger, U., Carslaw, K. Kulmala, M.: The effect of acid-base clustering and ions on the S., Curtius, J., Hansel, A., Petäjä, T., Kulmala, M., Donahue, growth of atmospheric nano-particles, Nat. Commun., 7, 11594, N. M., and Worsnop, D. R.: Molecular understanding of atmo- doi:10.1038/ncomms11594, 2016. spheric particle formation from sulfuric acid and large oxidized McMurry, P., Ghimire, A., Ahn, H.-K., Sakurai, H., Moore, K., organic molecules, P. Natl. Acad. Sci. USA, 110, 17223–17228, Stolzenburg, M., and Smith, J.: Sampling Nanoparticles for doi:10.1073/pnas.1306973110, 2013. Chemical Analysis by Low Resolution Electrical Mobility Clas- Schobesberger, S., Franchin, A., Bianchi, F., Rondo, L., Duplissy, J., sification, Environ. Sci. Technol., 43, 4653–4658, 2009. Kürten, A., Ortega, I. K., Metzger, A., Schnitzhofer, R., Almeida, O’Dowd, C. D., Aalto, P., Hämeri, K., Kulmala, M., and Hoff- J., Amorim, A., Dommen, J., Dunne, E. M., Ehn, M., Gagné, S., mann, T.: Atmospheric particles from organic vapours, Nature, Ickes, L., Junninen, H., Hansel, A., Kerminen, V.-M., Kirkby, 416, 497–498, 2002. J., Kupc, A., Laaksonen, A., Lehtipalo, K., Mathot, S., Onnela, Ouyang, H., He, S., Larriba-Andaluz, C., and Hogan, C. J.: IMS- A., Petäjä, T., Riccobono, F., Santos, F. D., Sipilä, M., Tomé, MS and IMS-IMS investigation of the structure and stability A., Tsagkogeorgas, G., Viisanen, Y., Wagner, P. E., Wimmer, D., of dimethylamine-sulfuric acid nanoclusters, J. Phys. Chem. A, Curtius, J., Donahue, N. M., Baltensperger, U., Kulmala, M., and 119, 2026–2036, doi:10.1021/jp512645g, 2015. Worsnop, D. R.: On the composition of ammonia-sulfuric-acid

www.atmos-chem-phys.net/16/13601/2016/ Atmos. Chem. Phys., 16, 13601–13618, 2016 13618 M. J. Lawler et al.: Acidic nanoparticles at CLOUD

ion clusters during aerosol particle formation, Atmos. Chem. Wong, J. P. S., Lee, A. K. Y., and Abbatt, J. P. D.: Impacts of sul- Phys., 15, 55–78, doi:10.5194/acp-15-55-2015, 2015. fate seed acidity and water content on isoprene secondary organic Simon, M., Heinritzi, M., Herzog, S., Leiminger, M., Bianchi, F., aerosol formation., Environ. Sci. Technol., 49, 13215–13211, Praplan, A., Dommen, J., Curtius, J., and Kürten, A.: Detection doi:10.1021/acs.est.5b02686, 2015. of dimethylamine in the low pptv range using nitrate chemical Xue, J., Yuan, Z., Griffith, S. M., Yu, X., Lau, A. K. H., and Yu, J. Z.: ionization atmospheric pressure interface time-of-flight (CI-APi- Sulfate formation enhanced by a cocktail of high NOx, SO2, par- TOF) mass spectrometry, Atmos. Meas. Tech., 9, 2135–2145, ticulate matter and droplet pH during haze-fog events in megaci- doi:10.5194/amt-9-2135-2016, 2016. ties in China: an observation-based modeling investigation, Envi- Smith, J. N., Moore, K. F., McMurry, P. H., and Eisele, F. L.: ron. Sci. Technol., 50, 7325–7334, doi:10.1021/acs.est.6b00768, Atmospheric Measurements of Sub-20 nm Diameter Particle 2016. Chemical Composition by Thermal Desorption Chemical Ion- Yli-Juuti, T., Nieminen, T., Hirsikko, A., Aalto, P. P., Asmi, E., ization Mass Spectrometry, Aerosol Sci. Tech., 38, 100–110, Hõrrak, U., Manninen, H. E., Patokoski, J., Dal Maso, M., doi:10.1080/02786820490249036, 2004. Petäjä, T., Rinne, J., Kulmala, M., and Riipinen, I.: Growth Smith, J. N., Dunn, M. J., VanReken, T. M., Iida, K., Stolzen- rates of nucleation mode particles in Hyytiälä during 2003– burg, M. R., McMurry, P. H., and Huey, L. G.: Chemical com- 2009: variation with particle size, season, data analysis method position of atmospheric nanoparticles formed from nucleation and ambient conditions, Atmos. Chem. Phys., 11, 12865–12886, in Tecamac, Mexico: Evidence for an important role for organic doi:10.5194/acp-11-12865-2011, 2011. species in nanoparticle growth, Geophys. Res. Lett., 35, L04808, Zhang, S.-H., Akutsu, Y., Russell, L. M., Flagan, R. C., and Sein- doi:10.1029/2007GL032523, 2008. feld, J. H.: Radial Differential Mobility Analyzer, Aerosol Sci. Smith, J. N., Barsanti, K. C., Friedli, H. R., Ehn, M., Kulmala, M., Tech., 12, 357–372, 1995. Collins, D. R., Scheckman, J. H., Williams, B. J., and McMurry, Zhao, J., Ortega, J., Chen, M., McMurry, P. H., and Smith, J. N.: De- P. H.: Observations of aminium salts in atmospheric nanoparti- pendence of particle nucleation and growth on high-molecular- cles and possible climatic implications, P. Natl. Acad. Sci. USA, weight gas-phase products during ozonolysis of a-pinene, At- 107, 6634–6639, doi:10.1073/pnas.0912127107, 2010. mos. Chem. Phys., 13, 7631–7644, doi:10.5194/acp-13-7631- Wang, L., Khalizov, A. F., Zheng, J., Xu, W., Ma, Y., Lal, V., 2013, 2013. and Zhang, R.: Atmospheric nanoparticles formed from het- Zollner, J. H., Glasoe, W. A., Panta, B., Carlson, K. K., McMurry, erogeneous reactions of organics, Nat. Geosci., 3, 238–242, P. H., and Hanson, D. R.: Sulfuric acid nucleation: power de- doi:10.1038/ngeo778, 2010. pendencies, variation with relative humidity, and effect of bases, Wang, M. and Penner, J. E.: Aerosol indirect forcing in a global Atmos. Chem. Phys., 12, 4399–4411, doi:10.5194/acp-12-4399- model with particle nucleation, Atmos. Chem. Phys., 9, 239–260, 2012, 2012. doi:10.5194/acp-9-239-2009, 2009. Weber, R. J., Marti, J. J., Mcmurry, P. H., Tanner, D. J., and Jeffer- son, A.: Measurements of new particle formation and ultrafine particle growth rates at a clean continental site, J. Geophys. Res., 102, 4375–4385, 1997.

Atmos. Chem. Phys., 16, 13601–13618, 2016 www.atmos-chem-phys.net/16/13601/2016/